Urban community low-carbon updating multi-objective optimization processing method and system

Abstract

The invention discloses an urban community low-carbon updating multi-objective optimization processing method and system, and belongs to the technical field of big data. Comprising the following steps that data are obtained and processed, the obtained data comprise geospatial data and population data, and a multi-objective optimized agent database and a spatial database are constructed; according to the constructed agent database, creating a simulation model of multi-objective optimization; according to the simulation model and in combination with external influence parameters, designing an update process mechanism, formulating update scheme selection scenes and conditions and intelligent agent interaction, and generating an optimal update strategy; and extracting and visually updating a multi-objective optimization result of the strategy. Compared with the prior art, the method has the advantages that multi-target optimization based on multi-agent interaction and space-time visualization of the updating process are achieved, interaction inside updating subjects such as residents and the government and interaction between the subjects are simulated, and multi-target optimization such as the minimum cost, the maximum carbon reduction amount and the maximum resident satisfaction degree is completed.

Description

Technical Field

[0001] The present invention relates to the field of big data technology, and more specifically, to a multi-objective optimization processing method and system for low-carbon renewal of urban communities. Background Art

[0002] As a basic component of the urban structure, communities account for a certain proportion of the city's overall carbon emissions. Community renewal, also known as community transformation, is the improvement and renewal of various problems or needs in the community, aiming to improve the quality of the living environment, promote harmonious social development, and meet the needs and expectations of residents.

[0003] However, current community renewal actions are mostly limited to economic perspectives, or only focus on single dimensions such as society, ecology, and culture, lacking comprehensive considerations on multi-objective optimization methods. Multi-objective optimization methods for community low-carbon renewal are relatively insufficient, and measurement is mostly based on a single goal, ignoring the complex adaptive system characteristics of urban renewal based on multi-dimensional value reorganization of different renewal subject combinations, capital combinations, and factor combinations. Therefore, how to coordinate multiple subjects, coordinate renewal costs and long-term carbon reduction goals and tasks, and how to scientifically complete planning decisions with public participation have become important issues in current community low-carbon renewal. With the continuous changes in information technology, geographic information systems and artificial intelligence provide new ideas and analysis methods for planning decisions. Multi-objective optimization algorithms can balance multiple conflicting goals and find the relatively optimal solution, and are widely used in planning decision-making research.

[0004] At present, although multi-objective optimization based on genetic algorithms has been gradually applied in many fields, most of them are based on programming software platforms such as MATLAB, Python, and Grasshopper. Although they can realize the visualization of multi-objective optimization result data based on static data, they cannot interact with resident intelligent agents and dynamically display the residents' update plan selection in time and space. Although Anylogic software can display the update status of residents under different parameters in real time through multi-agent simulation, its internal optimization module only supports single-objective optimization and cannot realize the effective integration of multi-objective optimization and complex interactive dynamic simulation of intelligent agents. Summary of the invention

[0005] 1. Technical issues to be solved

[0006] In view of the problems in the prior art that the current community low-carbon renewal measurement target is single and the dynamic update of the implementation subject's needs is insufficiently considered, the present invention provides a multi-objective optimization processing method and system for urban community low-carbon renewal, which can realize the dynamic demand analysis of multiple renewal subjects based on spatial digitalization construction, simulate the interaction of intelligent entities such as residents and governments, and take maximizing carbon reduction effects, minimizing renewal investment costs and maximizing resident satisfaction as renewal goals, and perform multi-objective iterative optimization of quantitative indicators in the renewal strategy to support low-carbon-oriented planning decisions in adaptive renewal.

[0007] 2. Technical solution

[0008] The purpose of the present invention is achieved through the following technical solutions.

[0009] The content of this application is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this application is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.

[0010] Some embodiments of the present application propose a multi-objective optimization processing method and system for low-carbon renewal of urban communities to solve the technical problems mentioned in the above background technology section.

[0011] As the first aspect of the present application, some embodiments of the present application provide a multi-objective optimization processing method for low-carbon renewal of urban communities, including the following steps: acquiring and processing data, the acquired data including geospatial data and population data, constructing an intelligent agent database and a spatial database for multi-objective optimization; creating a simulation model capable of multi-objective optimization based on the constructed intelligent agent database; based on the simulation model and in combination with external influencing parameters, generating an optimal update strategy by designing an update process mechanism, formulating update plans, selecting scenarios and conditions, and interacting with intelligent agents; extracting and visualizing the multi-objective optimization results of the optimal update strategy.

[0012] Furthermore, processing the data includes obtaining original carbon emission data of buildings, obtaining preset carbon emission reduction data, creating update plans, and constructing an intelligent database; the original carbon emission data of buildings includes original carbon emission data of residential buildings and original carbon emission data of non-residential buildings.

[0013] Furthermore, the geographic spatial data includes block terrain data, plot boundary data, building data, green space data, water system data and road network data; the building data includes building outline data, building longitude and latitude data, building plot number data and building attribute data.

[0014] Furthermore, the spatial database is composed of block terrain data, plot boundary data, building outline data, green space data, water system data and road network data.

[0015] Furthermore, in the process of constructing the intelligent agent database, the building and the residents corresponding to the building are regarded as a sample to construct a community full sample data set; the samples in the community full sample data set are sampled to obtain a sampled sample set, and the resident data of the sampled samples in the sampled sample set are collected; data enhancement is performed based on the building attribute data to obtain the resident data of all samples in the community full sample data set.

[0016] Furthermore, a first data table is established based on the building attribute data in the full sample data set of the community, and a second data table is established based on the building attribute data in the sampling data set. By finding the most similar row in the second data table to the first data table, the sample that is closest to the unsampled sample in terms of building attributes is found in the sampled sample.

[0017] Furthermore, in the search process, the Euclidean distance is used as the similarity metric to obtain the resident data of all samples in the community full sample data set; assuming that the characteristics of each sample are represented by a vector, the characteristic vector of the unsampled sample i is represented by X i =(z i1 , z i2 , …, z in ), the building attribute data of the unsampled sample i is represented by z ik ; The feature vector of sample j is represented by X j =(z j1 , z j2 , …, z jn ), the building attribute data of sample j is represented by z jk ; k is an index used to traverse the dimension of each vector. k is a natural number. The expression for calculating the Euclidean distance is as follows:

[0018]

[0019] Where d represents the Euclidean distance between the unsampled sample i and the sampled sample j, and n represents the dimension of the vector, that is, the number of building attribute variables.

[0020] Furthermore, the process of creating the simulation model includes: adding agents according to the agent database, and embedding NSGA-Ⅱ; the agents include a first agent group, a second agent group, and a NSGA-Ⅱ single agent.

[0021] Furthermore, the implementation process of NSGA-Ⅱ includes: initializing the population and generating random individuals; performing non-dominated sorting and calculating crowding distance; generating offspring populations through selection, crossover and mutation; merging the parent and offspring populations, performing non-dominated sorting and crowding distance calculation again to generate a new population; obtaining the Pareto optimal solution set through iterative cycles of selecting parents, performing crossover and mutation, and updating the population.

[0022] As the second aspect of the present application, some embodiments of the present application provide a system based on the above-mentioned processing method for multi-objective optimization of low-carbon renewal of urban communities, including a data acquisition module: acquiring and processing data, the acquired data includes geospatial data and population data, and constructing an intelligent agent database and a spatial database for multi-objective optimization; a model construction module: creating a simulation model capable of multi-objective optimization based on the constructed intelligent agent database; a strategy optimization module: based on the simulation model and in combination with external influencing parameters, by designing an update process mechanism, formulating update plans, selecting scenarios and conditions, and intelligent agent interactions, to generate an optimal update strategy; a result output module: extracting and visualizing the multi-objective optimization results of the optimal update strategy.

[0023] 3. Beneficial effects

[0024] Compared with the existing technology, the advantages of the present invention are: this scheme can realize multi-objective optimization based on multi-agent interaction and spatiotemporal visualization of the renewal process, simulate the interaction within and between renewal-related subjects such as residents and governments, complete multi-objective optimization such as minimum cost, maximum carbon reduction, and maximum resident satisfaction, and realize the visualization of long-term community low-carbon renewal progress based on spatial digital construction, provide scientific and quantitative indicator recommendations for the next renewal action, assist renewal subjects in evaluating investment costs, carbon reduction effects and renewal timing, and support low-carbon adaptive renewal of urban communities. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a flow chart of a multi-objective optimization processing method for low-carbon renewal of urban communities in one embodiment of the present invention;

[0026] Figure 2 It is a flowchart of the NSGA-II algorithm implementation in one embodiment of the present invention;

[0027] Figure 3 It is a logic diagram of the low-carbon renewal process of urban communities in one embodiment of the present invention;

[0028] Figure 4 It is a flowchart for realizing the visualization of the update status of the resident intelligent body in one embodiment of the present invention;

[0029] Figure 5A time line graph of the positive impact parameters of the community low-carbon renewal program in one embodiment of the present invention;

[0030] Figure 6 A time line chart showing the implementation ratio of the update scheme for low-carbon update of a community in one embodiment of the present invention;

[0031] Figure 7 A time line chart of the renewal investment value of the community low-carbon renewal in the current year in one embodiment of the present invention;

[0032] Figure 8 A time line chart of the cumulative updated carbon reduction amount of community low-carbon renewal in one embodiment of the present invention;

[0033] Fig. 9 A time line graph of feedback parameters of community low-carbon renewal in one embodiment of the present invention;

[0034] Fig.10 This is a timing diagram of update selection transition of an update object in one embodiment of the present invention. DETAILED DESCRIPTION

[0035] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0036] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0037] according to Figures 1 to 10 As shown, a multi-objective optimization processing method for low-carbon renewal of urban communities of the present invention includes the following steps: acquiring and processing data, constructing a multi-objective optimization intelligent agent database and a spatial database; creating a simulation model capable of multi-objective optimization based on the constructed intelligent agent database; deepening the design of the simulation model, combining the simulation model with external influencing parameters, and generating an optimal renewal strategy; extracting multi-objective optimization results, and visualizing the community low-carbon renewal process based on the spatial database.

[0038] like Figure 1 As shown, in a specific embodiment, the specific steps of the multi-objective optimization processing method for low-carbon renewal of urban communities include:

[0039] S1. Build a database

[0040] Acquire and process data, and build a multi-objective optimization agent database and a spatial database.

[0041] Specifically, the data acquired include geospatial data and population data.

[0042] Geographical data includes block terrain data, plot boundary data, building data, green space data, water system data and road network data.

[0043] The building data includes building outline data, building longitude and latitude data, building plot number data and building attribute data. Building attribute data includes building ownership, building style, building quality, building heritage protection strategy type, building function, building number of floors, number of residents, building perimeter and building area.

[0044] Specifically, the spatial database consists of block terrain data, plot boundary data, building outline data, green space data, water system data and road network data.

[0045] In a specific embodiment, the collected data is processed to obtain carbon emission data and construct an agent database, and the process is as follows:

[0046] (1) Obtaining raw building carbon emission data

[0047] Based on the geographic space data, the building data within the corresponding range of the community is locked, and the original carbon emission data of the building is calculated according to the building data and the population data corresponding to the building data.

[0048] Specifically, buildings include residential buildings and non-residential buildings, and the raw carbon emission data of buildings include the raw carbon emission data of residential buildings and the raw carbon emission data of non-residential buildings.

[0049] The original carbon emission data of residential buildings includes traffic carbon emission data and life carbon emission data, and the original carbon emission data of residential buildings is calculated based on population data. Among them, the expression for obtaining traffic carbon emission data is:

[0050] C t =∑D u *M u

[0051] In the formula, C t is the total carbon emissions from community transportation, in kg / year; u is the type of transportation, D u represents the annual travel distance under transportation mode u, M u is the carbon emission coefficient corresponding to different modes of transportation.

[0052] In this embodiment, the carbon emission coefficient values ​​corresponding to different transportation modes are shown in Table 1.

[0053] Table 1 Carbon emission coefficient of transportation carbon emission data

[0054] Transportation Carbon emission factor walk 0 bike 0 Battery car 0.0696 motorcycle 0.1136 Private Car 0.1786 taxi 0.1786 bus 0.0738 Company shuttle bus 0.0738

[0055] The living carbon emission data includes the carbon emission data of water use, wastewater use, electricity use, gas use and garbage disposal based on population data. The building and the residents corresponding to the building are regarded as a sample i. The expression for obtaining the living carbon emission data is:

[0056] C r =∑[R i *(W i *M W +Wi i *P w *M N +E i *M E +Q i *M Q +F i *M F )];

[0057] In the formula, C r is the total carbon emissions of community life, in kg / year; R i represents the total number of residents obtained from population data, W i It indicates the per capita water consumption of residential buildings, in kg / (person·year); M W is the carbon emission coefficient of water use, P W It represents the water pollution coefficient, M N Represents the carbon emission coefficient of wastewater, E i It represents the per capita electricity consumption of residential buildings, in kWh / (person / year); M E Indicates the carbon emission coefficient of electricity consumption, Q i Indicates the per capita gas consumption of residential buildings, in m 3 / (person-year); M Q Indicates the carbon emission coefficient of gas consumption, F i It indicates the amount of garbage generated per capita in residential buildings, in kg / (person-year); M F Represents the carbon emission coefficient of waste disposal.

[0058] Based on the acquired traffic carbon emission data and life carbon emission data, the total community traffic carbon emission and the total community life carbon emission are added together to obtain the total original carbon emission of all residential buildings. The expression for calculating the original carbon emission data of residential buildings based on the total original carbon emission of all residential buildings is:

[0059]

[0060] In the formula, CarbonInitial r Represents the original carbon emission data of residential buildings, in kg / year; S a Indicates the building area; S b It represents the per capita living area.

[0061] Specifically, for non-residential buildings, the expression for obtaining the original carbon emission data of non-residential buildings is:

[0062] CarbonInitial o =S a *(W S *(M w +P W *M N )+E s *M E +F S *M F )

[0063] In the formula, CarbonInitial o Represents the original carbon emission data of non-residential buildings, in kg / year; W S Indicates the water consumption of non-residential buildings, in kg / (㎡·year); E s Indicates the electricity consumption of non-residential buildings, in kWh / (㎡·year); F S It indicates the discharge of non-residential building waste, in kg / (㎡·year); M W is the carbon emission coefficient of water use, P W It represents the water pollution coefficient, M N Represents the carbon emission coefficient of wastewater, M E Represents the carbon emission coefficient of electricity consumption, M Q Indicates the carbon emission coefficient of gas consumption, M F Represents the carbon emission coefficient of waste disposal.

[0064] In this embodiment, the carbon emission coefficient of water use, the carbon emission coefficient of wastewater, the carbon emission coefficient of electricity use, the carbon emission coefficient of gas use, and the carbon emission coefficient of garbage disposal are shown in Table 2.

[0065] Table 2 Carbon emission coefficient of living carbon emission data

[0066] Carbon emission factor Value <![CDATA[M W ]]> 0.0003 <![CDATA[P W ]]> 0.8 <![CDATA[M n ]]> 0.00025 <![CDATA[M E ]]> 0.9044 <![CDATA[M Q ]]> 2.9538 <![CDATA[M F ]]> 0.3

[0067] Through the above steps, it is possible to obtain the original carbon emission data of the building.

[0068] (2) Create an update plan

[0069] Based on the analysis of carbon emission data, geographic data and building attribute data, an update plan is created, and the carbon reduction amount and input value of the update plan are obtained.

[0070] Specifically, the update scheme includes a low-amplitude update scheme, a medium-amplitude update scheme, and a high-amplitude update scheme. Set l to indicate the type of update scheme, l = 1 indicates selecting a low-amplitude update scheme; l = 2 indicates selecting a medium-amplitude update scheme; l = 3 indicates selecting a high-amplitude update scheme; set p l is the update strategy, p1 represents the positive impact parameter of the low-amplitude update plan; p2 represents the positive impact parameter of the medium-amplitude update plan; p3 represents the positive impact parameter of the high-amplitude carbon reduction update plan.

[0071] The specific process of obtaining the carbon reduction and input value of the updated solution is as follows:

[0072]

[0073] In the formula, CarbonReduce l represents the annual carbon reduction of the household when the residents choose the l type renewal plan, h represents the carbon reduction measure that measures the carbon reduction effect in units of people, and CMeasure h represents the per capita carbon reduction of carbon reduction measure h, v represents the carbon reduction measure whose carbon reduction effect is measured in units of area, and CMeasure v represents the carbon reduction per unit area of ​​carbon reduction measure v, S a Indicates the building area; S b It represents the per capita living area.

[0074]

[0075] In the formula, Investment l represents the renovation investment when residents choose type l renovation plan, e represents the carbon reduction measures that measure carbon reduction investment in units of people, and IMeasure e represents the per capita input value of carbon reduction measure h, f represents the carbon reduction measure that measures carbon reduction input in units of area, and IMeasure f represents the unit area input value of carbon reduction measure f, S a Indicates the building area; S b It represents the per capita living area.

[0076] (3) Building an agent database

[0077] Based on geographic space data, original building carbon emission data and population data, the building and the residents corresponding to the building are regarded as a sample to construct a community full sample data set; the samples in the community full sample data set are sampled to obtain a sample set, and the resident data of the sampled samples in the sample set is collected; data enhancement is performed based on the building attribute data to obtain the resident data of all samples in the community full sample data set. Based on the original building carbon emission data, building latitude and longitude data, building location block number data, carbon reduction and input value data of the update plan, and resident data, an intelligent agent database is constructed.

[0078] Specifically, the samples in the community full sample data set include building attribute data, building latitude and longitude data, building plot number data, building original carbon emission data, carbon reduction and input value of the update plan, and the data collected for the sampled samples refers to the resident data of the sampled samples. Data enhancement refers to the technology of expanding training data by using algorithms, and the method of expanding the training data set by using a small amount of data prior knowledge to generate more similar generated data. Through the process of data enhancement of the sampled samples, the resident data of all samples in the community full sample data set are obtained. This step can effectively expand the data set, propose remedial solutions for scenarios where full sample surveys cannot be achieved during the update practice, and minimize data errors.

[0079] In this embodiment, data enhancement adopts the Propensity Score Matching (PSM), which is a non-experimental attribution analysis method. It uses observable variables and logistic regression to find a group of comparable control groups (control groups). By calculating the propensity value and integrating multiple confounding variables, the difficulty of matching is greatly reduced. The PSM method uses the propensity score function to compress the information of the multidimensional vector into one dimension, and then matches according to the propensity score. In this way, the samples of the treatment group and the control group can be made as similar as possible under the given observable characteristic variables. The basic idea of ​​PSM includes: finding a sample of the control group (control group) so that it matches (similar) as much as possible with the samples of the treatment group (experimental group) in the values ​​of the observable variables. In this embodiment, the propensity score matching is implemented based on the MatchIt package and the dplyr package in the R language.

[0080] Specifically, a first data table is established based on the building attribute data in the community full sample data set, and a second data table is established based on the building attribute data in the sampled data set. By finding the most similar row in the second data table to the first data table, the sample closest to the unsampled sample in terms of building attributes is found in the sampled sample. More specifically, the Euclidean distance is used as the similarity metric in the search process. The smaller the Euclidean distance, the higher the similarity between the two samples in terms of building attributes, which effectively screens out the sampled sample closest to the unsampled sample in terms of building attributes, thereby obtaining the resident data of all samples in the community full sample data set.

[0081] In a specific embodiment, assuming that the feature of each sample is represented by a vector, the feature vector of the unsampled sample i is represented by X i =(z i1 , z i2 , …, z in ), the building attribute data of the unsampled sample i is represented by z ik ; The feature vector of sample j is represented by X j =(z j1 , z j2 , …, z jn ), the building attribute data of sample j is represented by z jk ; k is an index used to traverse the dimension of each vector, and k is a natural number. The expression for calculating the Euclidean distance is as follows:

[0082]

[0083] Where d represents the Euclidean distance between the unsampled sample i and the sampled sample j, and n represents the dimension of the vector, that is, the number of building attribute variables.

[0084] Traverse the unsampled sample data set: Take unsampled sample 1 as an example, calculate the Euclidean distance between this sample and each sampled sample, select the sampled sample with the smallest Euclidean distance, extract its serial number, and match the resident data corresponding to this sampled sample; the same applies to unsampled samples 2, 3, 4, etc. Find the serial number of the resident data and the corresponding resident data according to the serial number, export and save in csv format, which is the resident data of all samples in the community full sample data set.

[0085] An intelligent database is constructed based on the building’s latitude and longitude data, the building’s plot number data, the building’s original carbon emission data, the carbon reduction and input value of the update plan, and resident data.

[0086] S2. Create a simulation model

[0087] Based on the constructed intelligent agent database, a simulation model capable of multi-objective optimization is created.

[0088] Specifically, the problem to be optimized is modeled, and then a simulation model is created. According to the agent database constructed in step S1, agents are added, and a multi-objective optimization genetic algorithm is embedded to simulate the actions and decisions of multiple subjects in the update process, as well as iterative optimization experiments of update strategies, and the optimal update strategy is generated based on the influencing factors and update schemes. In this embodiment, the influencing factor can be a tendency of the update object to the update scheme, and the tendency to the update scheme includes tendency and non-inclination. The specific process is as follows:

[0089] (1) Establishing a problem model

[0090] Building a problem model includes defining genes, objective functions, and constraints. Among them, genes refer to the specific optimization objects, that is, the indicators of the update strategy: positive impact parameters.

[0091] First, the problem to be solved is modeled as an optimizable problem. In this embodiment, the problem to be solved is determined as: how to balance the feedback parameters, the input value required for the update scheme and the carbon reduction amount to achieve low-carbon update based on the data in the intelligent agent database? More specifically, the optimal update strategy is formed by locking the optimal update objects in different time periods and generating the current optimal carbon reduction amount at the same time.

[0092] Specifically, the feedback parameter may be a value reflecting the satisfaction of the update target with the update solution.

[0093] Set fitness as the variable to verify the optimization results; the input value of the current year is fitness1; the cumulative updated carbon reduction is fitness2.

[0094] Specifically, the objective function defined includes minimizing the input value of the current year and maximizing the cumulative updated carbon reduction. The expression of the objective function is as follows:

[0095]

[0096] In the formula, cost il Indicates the input value in the scenario where update object i selects solution l, carbonreduce l It represents the carbon reduction amount produced by updating the scheme l of object i.

[0097] Set the carbon reduction basic amount carbonTask as a constraint condition, and use the carbon reduction basic amount to indicate that the updated carbon reduction amount is the minimum value of fitness2. The carbon reduction basic amount expression is as follows:

[0098] carbonTask=carbonTotal*n+carbonreduce;

[0099] In the formula, carbonTotal represents the total carbon emissions of the community after the last update, n refers to the ratio of the minimum carbon reduction produced by this update to the total carbon emissions after the last update, and carbonreduce refers to the cumulative carbon reduction of the community after the last update.

[0100] (2) Embedding multi-objective optimization genetic algorithm

[0101] NSGA-Ⅱ is used as an embedded multi-objective optimization genetic algorithm. NSGA-Ⅱ (Non-dominated Sorting Genetic Algorithm Ⅱ) is a multi-objective optimization genetic algorithm used to solve optimization problems with multiple conflicting objective functions. In this embodiment, since the greater the carbon reduction, the higher the carbon reduction input value, in order to achieve a balance between carbon reduction and input value, NSGA-Ⅱ is selected to be embedded. The process of NSGA-Ⅱ includes: initializing the population, generating individual codes and evaluating the objective function values ​​of individuals; performing non-dominated sorting and crowding distance calculation on individuals in the initial population, and generating offspring populations through selection, crossover and mutation operations; after merging the parent population and the offspring population, performing non-dominated sorting and crowding distance calculation on individuals in the merged population again, and screening to generate a new population; repeating the above process until the preset termination condition is met. Through such a process, a set of optimal equilibrium solutions can be found between multiple conflicting objectives, namely, the Pareto optimal solution set.

[0102] Specifically, the preset termination condition may be reaching a maximum number of iterations.

[0103] The update strategy set according to the update scheme (including the low-amplitude update scheme, the medium-amplitude update scheme and the high-amplitude carbon reduction update scheme) is regarded as an individual, and the encoding data of each individual includes: gene, fitness, non-dominated level, number of dominated, crowding distance, dominating set and number of times dominated.

[0104] Embed NSGA-Ⅱ in the simulation model. The specific steps are as follows:

[0105] First, build a simulation model in AnyLogic and configure the model database, resources, environment base map and other related parameters of the simulation model. Other related parameters include space type, time unit, simulation memory, model time execution mode, model time execution ratio and running configuration memory.

[0106] In a specific embodiment, the spatial type is defined as GIS, urban blocks and building form files are added as environmental base maps, the time unit is set to month, the simulation memory is 50000Mb, the model time execution ratio is 1, and the running configuration memory is 16384Mb.

[0107] Secondly, constructing the intelligent agent includes constructing the first intelligent agent group, the second intelligent agent group and the NSGA-Ⅱ single intelligent agent. The parameter setting of the first intelligent agent group is completed by mapping the intelligent agent database, and the coding data of the update strategy is established in the second intelligent agent group.

[0108] In a specific embodiment, the first agent group is the resident agent group, and the second agent group is the update strategy agent group. The process of constructing the first agent group is: right-click the Main agent in the project list, create a new agent and name it Resident, use the table that records the update feature data as the basic database table, select all columns in the table as parameters of the first agent group Resident, perform parameter mapping inside Resident, and complete the creation of the first agent. Create new variables choice, dyy, zyy, gyy and satisfaction inside Resident, which are used to record the update object's choice of update plan, the tendency to choose a low-amplitude update plan, the tendency to choose a medium-amplitude update plan, the tendency to choose a high-amplitude update plan, and feedback parameters. The feedback parameter is a value that reflects the update object's satisfaction with the update plan.

[0109] The process of building the second agent is as follows: right-click the Main agent in the project list, create a new agent and name it Zc; double-click the second agent group Zc in the project, add agent components through the panel, copy the first agent group Resident and paste it into the second agent group Zc, except for the coding information, to record the different feedback of the updated object in different positive impact parameter combinations. l is point data, and the positive impact parameter of interest to the update object is interval data. Therefore, in order to facilitate subsequent comparison, a conversion variable Type is added inside the second agent group Zc l (l=1,2,3). In this embodiment, the conversion variable Type l The specific values ​​of and their corresponding positive impact parameters are shown in Table 3.

[0110] Table 3 Conversion variable Type l The specific value of and its corresponding positive impact parameter

[0111] Update strategy <![CDATA[Conversion variable Type l <!-- 8 -->]]> <![CDATA[p l =0]]> 1 <![CDATA[0.01≤p l ≤0.25]]> 2 <![CDATA[0.26≤p l ≤0.50]]> 3 <![CDATA[0.51≤p l ≤0.75]]> 4 <![CDATA[0.76≤p l ≤1]]> 5

[0112] After the first and second agent groups are built, a new NSGA-Ⅱ single agent is created. Based on the state diagram module in the AnyLogic agent panel, NSGA-Ⅱ is split into different "states". Different "states" are connected by "transitions". The specific process is implemented by entering the action code internally or calling a specified function to determine whether to enter the next state. The state is the specific situation or mode that the agent is in at a certain moment. In the NSGA-Ⅱ single agent, each state can represent a stage or step of the algorithm, such as initialization of the population, non-dominated sorting, crowding calculation, selection, crossover, mutation, etc. Transition refers to the process or event of the agent moving from the current state to the next state, indicating the transition of the algorithm from one stage to another, such as the transition from the initialization state to the fast non-dominated sorting state.

[0113] like Figure 2 As shown in the figure, the implementation process of NSGA-Ⅱ in the simulation model is as follows:

[0114] 1) Initialize the population and generate random individuals:

[0115] Set the population size, generate a random group of individuals and calculate the fitness value, select individuals that meet the basic carbon reduction carbonTask and generate the initial population, set the size of the target Pareto optimal solution set and the maximum number of iterations and other parameters. The update strategy p l It is a random number with two decimal places and a value between 0 and 1.

[0116] 2) Non-dominated sorting and calculation of crowding distance:

[0117] Calculate the dominance relationship and the number of dominated for each individual; group the individuals according to the dominance relationship; repeatedly extract non-dominated solutions from ungrouped individuals until all individuals are grouped.

[0118] Specifically, the core of NSGA-Ⅱ is that it uses a fast non-dominated sorting method to sort individuals in the population. This sorting method is based on the dominance relationship between individuals. In the optimization problem, if an individual is not worse than another individual in all objective functions and is better than another individual in at least one objective function, then the individual is said to dominate another individual.

[0119] In a specific embodiment, the dominates() function and the nonDominatedSorting() function are customized for non-dominated sorting. The dominates() function is used to determine the dominance relationship between two individuals. If individual p dominates individual q, it returns true; otherwise, it returns false. The nonDominatedSorting() function stratifies the individuals and adds them to different fronts according to the dominance relationship between the individuals. The smaller the front value, the higher the individuals in the front are ranked in the population (that is, they are non-dominated solutions or are less dominated by other solutions).

[0120] To calculate the crowding distance, individuals need to be sorted according to the objective function value. The crowding distance of the boundary individuals is infinite, while the crowding distance of the middle individuals is calculated as follows:

[0121]

[0122] Among them, i d represents the crowding distance of individual i in the population, represents the function value of the g-th objective function of the i+1-th individual, where i and g are both natural numbers. represents the function value of the g-th objective function of the i-1th individual, maxf g represents the maximum value of the g-th objective function of individual i, minf g represents the minimum value of the g-th objective function of individual i, and m represents the total number of objective functions.

[0123] In this embodiment, the calculation of the crowding distance of individuals in a population is implemented by the calculateCrowdingDistance() function, and the steps include: for each individual in the population, the crowding distance is initialized to 0; for each objective function, the crowding distance of the best performing (i.e., the smallest value) and worst performing (i.e., the largest value) individuals in the population are set to infinity; for non-boundary individuals in the population, their crowding distances on each objective function are calculated; these differences are added to obtain the crowding distance of the individual in a given target space; finally, the calculated crowding distance is returned to each individual.

[0124] 3) Generate offspring population through selection, crossover and mutation:

[0125] First, the binary tournament method is used to screen a specified number of parent individuals. Two individuals in the original population are randomly selected, and their non-dominated ranks are compared. The individual with the smaller rank value is selected as the parent individual. When the rank values ​​of the two individuals are the same, that is, they are at the same frontier, the individual with the larger crowding distance is selected as the parent individual.

[0126] In a specific embodiment, a specified number of parent individuals are screened and added to the parent population. The original population size is 100, the number of parent individuals is 15, and the parent population storing the parent individuals is a collection parents, and the collection type is ArrayList.

[0127] Secondly, pairing, crossover and mutation operations are performed. In the NSGA-II algorithm, commonly used crossover operation methods include simulated binary crossover, single-point crossover, multi-point crossover and uniform crossover, and mutation operation methods include polynomial mutation, Gaussian mutation, non-uniform mutation and uniform mutation.

[0128] Specifically, the gene sequence in this embodiment only involves a single variable type, so the uniform crossover and uniform mutation methods are selected, the gene mutation probability is set to 0.1, and the target offspring population size is 100. The specific process of performing pairing, crossover and mutation operations is as follows: randomly generate a specified number of pairs of individuals from the selected parent population. In this embodiment, a total of 15 pairs are generated and stored in the set pairs; uniform crossover is performed on the genes in the paired parent to generate a corresponding number of offspring individuals; uniform mutation is performed on each offspring individual according to the mutation probability, and offspring individuals that meet the basic amount of carbon reduction are screened to generate a specified size of offspring population, which is stored in the children set, and the collection type is ArrayList.

[0129] 4) Merge the parent and child populations, perform non-dominated sorting and crowding distance calculation again, and generate a new population:

[0130] The initial population and the offspring population are merged to generate a new population. The size of the new population is the sum of the number of individuals in the initial population and the offspring population. The fitness of the new population is calculated, and non-dominated sorting and crowding distance are performed to ensure the integrity of the individual encoding information in the merged population. Finally, according to the Pareto hierarchy of the individuals and the internal sorting of the hierarchy, a specified number of frontier individuals are selected to form a new population.

[0131] In a specific embodiment, the function merge() is used to merge populations, and the merged population is stored in the set merged_population, the set type is ArrayList, and the population size is 200; the method used to filter the first specified number of individuals in the population is the function regenerateChoose(targetNum), where targetNum represents the number of target individuals to be calculated and filtered, and the data type of targetNum is int. This step filters the first 100 individuals in the set merged_population to generate a new population.

[0132] 5) Iterate the loop to get the Pareto optimal solution set:

[0133] By continuously selecting parents, performing crossover and mutation, and updating the population until the maximum number of iterations is reached, after the iteration is completed, a specified number of frontier individuals are selected as the Pareto optimal solution set based on the Pareto hierarchy of individuals in the final population and the internal sorting of the hierarchy.

[0134] In a specific embodiment, the number of iterations is set to 10, and the individual ranked first in the first-layer frontier is selected as the Pareto optimal solution.

[0135] (3) Run the simulation model

[0136] After embedding the multi-objective optimization genetic algorithm, settings are made for the first agent group, the second agent group and the NSGA-Ⅱ single agent, including the agent time mode selection, the first update time or the recurrence time, and whether to create a data set for the dynamic variable; after the settings are completed, the final simulation model is obtained.

[0137] By running the simulation model, verify the console output information to obtain the running results of the simulation model.

[0138] In a specific embodiment, the operation result of the simulation model is verified by the traceln() function, and the operation result is evaluated after verification. The evaluation process includes determining whether the number of individuals and related coding values ​​in the Pareto optimal solution set reaches a preset value. During the evaluation process, the real-time value of the specified variable can be checked by using the traceln() function. Finally, the optimal update strategy is obtained through the constructed simulation model.

[0139] S3. Generate optimal update strategy

[0140] According to the constructed simulation model and combined with the influence of external influencing parameters on the update object during the update process, the optimal update strategy is generated.

[0141] Specifically, the external influence parameter may be a parameter reflecting the degree to which the update object that has not been updated is affected by the update object that has been updated. The specific process of generating the optimal update strategy is as follows:

[0142] (1) Design update process mechanism

[0143] Generate an initial update plan; execute NSGA-Ⅱ to generate the optimal strategy. Due to the influence of the positive influencing parameters, the influencing factors will change for the first time, and due to the influence of the external influencing parameters, the influencing factors will change for the second time. Therefore, according to the changes in the influencing factors, the multi-objective optimization strategy needs to be updated again, and the above process is iterated repeatedly until the target node is reached.

[0144] In this embodiment, the influence factor reflects the inclination of the update object to the update scheme; the external influence parameter reflects the degree to which the update object that has not been updated is affected by the updated update object; and the positive influence parameter reflects the incentive of the update strategy to implement the update scheme.

[0145] like Figure 3 As shown, this is the process of the update mechanism. In this embodiment, the update simulation model starts running in October 2024, generates an initial update plan based on the influencing factor, and obtains the optimal update strategy; after 6 months, due to the influence of the positive influencing parameter, the influencing factor changes for the first time, the inclination of the update object to the update plan changes, and the optimal update strategy of this round changes accordingly; after 12 months, due to the influence of the external influencing parameter, the influencing factor changes for the second time, the inclination of the update object to the update plan changes again, and the update plan changes again accordingly. Based on this situation, the final update strategy is obtained, and the iterative cycle is carried out according to this rule until 2030 and the update simulation model operation is stopped.

[0146] (2) Formulate update plan, select scenarios and conditions

[0147] The update scheme selection scenarios include the scenarios where the update strategy parameter interval is greater than the expected parameter interval of the update object, the update strategy parameter interval is equal to the expected parameter interval of the update object, or the scenario is affected by external influencing parameters. When the update object has the same inclination towards multiple update schemes and is inclined, the update scheme with a higher amplitude is selected. Based on the external influencing parameters, the probability probability of the update object's inclination towards the update scheme changes is calculated as follows:

[0148]

[0149] Among them, exist is used to indicate whether there is an update object updated by this update scheme. If exist = true, it means that there is an update object updated by this update scheme. If exist = false, it means that there is no update object updated by this update scheme. TQ refers to the degree of influence of the update object by the external influence parameter. MaxQS indicates the degree of influence of the update object with the largest external influence parameter value. The values ​​of TQ and maxQS are both integers between 1 and 5.

[0150] Based on geographic spatial data, building data and population data, since there are situations in which update objects tend to be updated but are not updated during the actual update process, a wait-and-see mechanism is established in this embodiment. The wait-and-see mechanism is: when more update objects are implemented for the update plan, the probability of the influencing factor being converted into an update selection will be higher, thereby improving the random probability setting of the influencing factor conversion tendency.

[0151] (3) Agent Interaction

[0152] The process of agent interaction is mainly achieved through direct message transmission, shared variables, using environmental objects as intermediaries, and event-triggered interaction. In this embodiment, agent interaction includes agent communication based on a message sending and receiving mechanism and agent interaction based on shared variables.

[0153] Specifically, agent communication is implemented based on the message sending and receiving mechanism. The function send() is used to send a message to a single agent. The function body is: public void send(java.lang.object msg, Agent dest). Among them, msg refers to a message, and dest refers to the target agent that receives the message.

[0154] Message reception is done through the state diagram inside the agent, and the trigger mode of the transition is set to message. When the trigger transition type is a specified message and the message is consistent with the content of the sent information, the connections between the agents can receive the message, trigger the transition, and complete the communication between the agents.

[0155] Specifically, agent interaction is achieved based on shared variables, that is, multiple agents can access and modify shared variables to interact. In this embodiment, the second agent group Zc can generate an update strategy based on the influencing factor and the selection of the update scheme, and the update object adjusts the influencing factor and the selection of the update scheme according to the update strategy. The influencing factor and the selection of the update scheme are shared variables that serve as feedback on the impact of the update strategy on the update object, and finally generate the optimal update strategy.

[0156] S4. Multi-objective optimization result extraction and update process visualization

[0157] After completing the construction of the update simulation model, in order to better display the multi-objective optimization results under the optimal update strategy, the environmental base map setting of the simulation model is improved based on the spatial database, and the optimal update object in different time periods is locked. First, the operation results of the update simulation model are extracted to obtain the indicator data; secondly, the update process is dynamically visualized in time and space through intelligent body animation and adding analysis charts.

[0158] Specifically, the indicator data include the positive impact parameters of each renewal scheme, the renewal ratio of each renewal scheme, the investment value of the current year, the cumulative renewal carbon reduction, and the feedback parameters.

[0159] The update of the indicator data is synchronized with the change of the update strategy adjustment. In this embodiment, a time line graph is used for chart visualization, and the time window of the time line graph is set to 6 years. Figures 5 to 9As shown, it is a time line chart of relevant indicator data.

[0160] Update rate for each update scheme l The calculation formula is as follows:

[0161]

[0162] Among them, count l represents the total number of update objects that choose scheme l for update in the community, R i Represents the total number of residents.

[0163] In this embodiment, the update object is the residents corresponding to the building, and the calculation formula of the feedback parameter satisfaction is as follows:

[0164]

[0165] Where level is the level value set according to the currently selected update scheme, level = 0, 1, 2, 3;

[0166] When level=0, the current selection is not updated;

[0167] When level=1, the low-amplitude update scheme is currently selected;

[0168] When level = 2, it corresponds to the current selection of the medium-amplitude update scheme;

[0169] When level=3, it corresponds to the current selection of high-amplitude update scheme;

[0170] Where m refers to the matching degree of the update strategy interval relative to the expected parameter interval of the update object, m = 1, 2, 3;

[0171] When m=1, it means that the update strategy interval is lower than the expected parameter interval of the update object;

[0172] When m=2, it means that the update strategy interval is equal to the expected parameter interval of the update object;

[0173] When m=3, it means that the update strategy interval is higher than the expected parameter interval of the update object; R i Represents the total number of residents.

[0174] Spatial visualization is based on the selection of update schemes of the update object at different time periods. The animation effect of the first agent changes with the update status of the update object. The function setFillColor() of the AnyLogic simulation model is used to define the fill color of the graphic animation. The function body is: Public void setFillColor(java.awt.ColorfillColor). Among them, fillColor refers to the new fill color.

[0175] like Figure 4 As shown in Figure 1, the implementation procedure of the visualization of the resident agent update status is shown. Fig.10 As shown, the changes in the selection of update schemes for update objects over time are demonstrated.

[0176] As shown in Table 4, the intelligent agent animation filling color corresponding to the update scheme selected by the update object in this embodiment.

[0177] Table 4: The fill color of the intelligent agent animation corresponding to the resident update selection

[0178] Update plan selection Agent animation fill color Unselected initial state white Not updating gainsboro Low-impact update plan yellowGreen Medium-scale update plan green High-level update plan darkGreen

[0179] In a specific embodiment, a multi-objective optimization processing system for low-carbon renewal of urban communities includes a data acquisition module: collecting and processing data, constructing a multi-objective optimization intelligent agent database and a spatial database; a model construction module: creating a simulation model capable of multi-objective optimization based on the constructed intelligent agent database; a strategy optimization module: deepening the simulation model design, combining the simulation model with external influencing parameters, and generating an optimal renewal strategy; a result output module: extracting multi-objective optimization results, and visualizing the community low-carbon renewal process based on the spatial database.

[0180] The multi-objective optimization processing method and system for low-carbon renewal of urban communities of the present invention, through spatial digitalization construction and dynamic demand analysis of multiple renewal subjects, simulates the interaction of intelligent entities such as residents and the government, and takes maximizing carbon reduction effects, minimizing renewal investment costs and maximizing resident satisfaction as renewal goals, and realizes multi-objective iterative optimization of quantitative indicators in renewal strategies to support low-carbon-oriented planning decisions in adaptive renewal.

[0181] The above schematically describes the invention and its implementation methods, which is not restrictive. Without departing from the spirit or basic features of the invention, the invention can be implemented in other specific forms. What is shown in the accompanying drawings is only one of the implementation methods of the invention. The actual structure is not limited to this, and any figure mark in the claims should not limit the claims involved. Therefore, if a person of ordinary skill in the art is inspired by it, without departing from the purpose of the invention, a structural method and an embodiment similar to the technical solution are designed without creativity, which should all fall within the scope of protection of this patent. In addition, the word "including" does not exclude other elements or steps, and the word "one" before the element does not exclude the inclusion of "multiple" elements. The multiple elements stated in the product claim can also be implemented by one element through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any specific order.

Claims

1. A multi-objective optimization processing method for low-carbon renewal of urban communities, comprising the following steps: Acquire and process data, including geographic spatial data and population data, and build a multi-objective optimization agent database and spatial database; Based on the constructed agent database, a simulation model capable of multi-objective optimization is created; According to the simulation model and combined with external influencing parameters, the optimal update strategy is generated by designing the update process mechanism, formulating the update plan selection scenarios and conditions, and the intelligent agent interaction; Extract and visualize the multi-objective optimization results of the update policy.

2. The method for multi-objective optimization of low-carbon renewal of urban communities according to claim 1 is characterized by: Data processing includes obtaining the original carbon emission data of the building, obtaining the preset carbon emission reduction data, creating an update plan, and building an intelligent database; The raw carbon emission data of buildings include the raw carbon emission data of residential buildings and the raw carbon emission data of non-residential buildings.

3. The method for multi-objective optimization of low-carbon renewal of urban communities according to claim 1 is characterized by: Geographical data include block terrain data, plot boundary data, building data, green space data, water system data and road network data; building data include building outline data, building longitude and latitude data, building plot number data and building attribute data.

4. The method for multi-objective optimization of low-carbon renewal of urban communities according to claim 3 is characterized by: The spatial database is composed of block terrain data, plot boundary data, building outline data, green space data, water system data and road network data.

5. The method for multi-objective optimization of low-carbon renewal of urban communities according to claim 1 is characterized by: In the process of constructing the intelligent agent database, the building and the residents corresponding to the building are regarded as a sample to construct a community full sample data set; the samples in the community full sample data set are sampled to obtain a sample set, and the resident data of the sampled samples in the sample set are collected; Data enhancement is performed based on building attribute data to obtain resident data of all samples in the community full sample dataset.

6. The method for multi-objective optimization of low-carbon renewal of urban communities according to claim 5 is characterized by: A first data table is established based on the building attribute data in the community full sample data set, and a second data table is established based on the building attribute data in the sampling data set. By finding the most similar row in the second data table to the first data table, the sample that is closest to the unsampled sample in terms of building attributes is found in the sampled sample.

7. The method for multi-objective optimization of low-carbon renewal of urban communities according to claim 6 is characterized by: In the search process, Euclidean distance is used as the similarity metric to obtain the resident data of all samples in the community full sample data set; assuming that the characteristics of each sample are represented by a vector, the characteristic vector of the unsampled sample i is represented by X i =(z i1 , z i2 , …, z in ), the building attribute data of the unsampled sample i is represented by z ik ; The feature vector of sample j is represented by X j =(z j1 , z j2 , …, z jn ), the building attribute data of sample j is represented by z jk ; k is an index used to traverse the dimension of each vector. k is a natural number. The expression for calculating the Euclidean distance is as follows: Where d represents the Euclidean distance between the unsampled sample i and the sampled sample j, and n represents the dimension of the vector, that is, the number of building attribute variables.

8. The method for multi-objective optimization of low-carbon renewal of urban communities according to claim 1 is characterized by: The process of creating a simulation model includes: adding agents according to the agent database, and embedding NSGA-Ⅱ; the agents include the first agent group, the second agent group and the NSGA-Ⅱ single agent.

9. The method for multi-objective optimization of low-carbon renewal of urban communities according to claim 1 is characterized by: The implementation process of NSGA-Ⅱ includes: initializing the population and generating random individuals; performing non-dominated sorting and calculating crowding distance; generating offspring populations through selection, crossover and mutation; merging the parent and offspring populations, performing non-dominated sorting and crowding distance calculation again to generate a new population; obtaining the Pareto optimal solution set through iterative cycles of selecting parents, performing crossover and mutation, and updating the population.

10. A system based on the multi-objective optimization processing method for low-carbon renewal of urban communities according to any one of claims 1 to 9, characterized in that: It includes a data acquisition module: acquiring and processing data, including geographic spatial data and population data, and building a multi-objective optimized intelligent agent database and a spatial database; Model building module: Create a simulation model capable of multi-objective optimization based on the constructed intelligent agent database; Strategy optimization module: Based on the simulation model and combined with external influencing parameters, the optimal update strategy is generated by designing the update process mechanism, formulating the update plan selection scenario and conditions, and the intelligent agent interaction; Result output module: extracts and visualizes the multi-objective optimization results of the update strategy.

Images